Behavioral economics and facebook conspiracy theories
Imminently confronting its own mortality, the advertising industry as we know it has been looking to behavioral economics to keep food on the table. Behavioral economics is cool and scary. Daniel Kahneman’s Thinking, Fast and Slow is the go-to text on the issue; it summarizes the past 50 years of psychology research that created the discipline and won the author his Nobel prize in Economics. If you’ve read the book, skip the next three paragraphs.
In a nutshell, the book explains how we, as thinking folk, have two systems at our disposal as we go about trying to understand the world. System 1 is what generally gets us through life, it deals with things very quickly. When we have a casual conversation, estimate how far away an object is, or drive a car, that’s System 1 doing its thing. For more complex problems the higher level but very lazy System 2 kicks in, like trying to multiply 231 times 371, or actively fact checking a claim.
System 1 is great at what it does, but it uses a number of short-hand tricks that sometimes yield bad outcomes. For instance, System 1 is awful at statistics, so estimations it makes about the likelihood of something happening are usually way off. Much of Thinking, Fast and Slow is a chronicle of these accidental misfires that we make in our thinking, and the situations that bring them about. When advertisers look to behavioral economics to save the day, it’s my opinion that they’re looking for ways to exploit these misfires and make people more likely to buy.
System 2 is also good at what it does, but with practice. When it’s called to action, it deals with the world in logical terms. It tends to be very lazy. In reading the above 231 times 371, I’m sure you didn’t bother to actually calculate the result (I didn’t). If you go back and try to do the multiplication, you’ll feel the blood flow to your brain as you develop a strategy to produce the result. This is called cognitive strain. Whenever you feel it, System 2 is dealing with the world for you.
This brings me to facebook Timeline. Try scanning someone’s Timeline. It’s a very unpleasant experience. When information is organized in a list, it’s trivially easy to scan it, but with Timeline your eye has to dart around and try to combine the layout into an understanding of what the person’s been up to. It induces cognitive strain and brings System 2 online.
There was an interesting post a few weeks ago about how Timeline’s core goal is to reconceptualize brands in the eyes of users, and integrate them into the stream of data we see. When we usually see ads, System 1 processes and largely ignores them. But if Timeline causes System 2 to come online, then displays ads to it, facebook will be changing the way our brains process advertising. Timeline also makes branded posts (ads) look nearly identical to the actual content we’re on facebook to see, so it follows that they’d be processed similarly.
Is it as effective? More effective? Less effective? As end users, we lack the data to test it, but facebook does not. Some A/B testing and trading of IP addresses with online stores and it would be trivially easy to tell if these ads are more effective than the usual ones. With 2011 revenues of about $3.3 billion spread across 800,000,000 users, they’re making about $4 a user a year. That’s pathetic given how much the site is used. By contrast, the movie industry, despite trying to weather some radical disruption, collects $10-15 for 90 minutes of entertainment. (Different business models, I recognize.)
This all reminds of a recent quote about the state of the world. Paraphrasing, “the smartest people in the world are working hard to come up with ways to get you to click on ads.” (Edit: source) And it’s absolutely true, Google, Facebook, Microsoft and the like are brain trusts. People recognize that advertising is about psychology, and the bible of psychological decision making was just published. Why wouldn’t these smart people try to use these new tools to make more money? Timeline is just awful, but these people are too smart and skilled to make something so bad accidentally.
It’s on a par with Zynga’s behavior, but if anything that just makes it more facebook-like, not less.
I have not seen ads within the profile timeline. Are they already active?
In Firefox 9.0.1 with JS turned on, the lower-case ‘r’ character in your font sits lower than the line of the rest of the text. Not much, but just enough to bother me and lead me to comment on it.
Every single time Facebook releases a new feature, they’re criticized for a host of reasons (too commercial, privacy-problems, bad interface, I just hate it, etc) and yet most people generally find that they appreciate the feature after using it for a while. I don’t doubt that Facebook is self-interested. They have every right to create features that give them more page views, that give them more information for advertising, etc. But there’s no nefarious conspiracy outside of the really smart engineers at Facebook trying to experiment with some cool new ideas that benefit their company that they think people will use. I also think companies will benefit from the Timeline because it should enable them to use their creativity to hit people emotionally and convert that into sales. Whether companies are using various content-marketing methods, Facebook ads directly, or third party types of services like the ecosystem at http://www.buyfacebookfansreviews.com for example, there’s a ton of ways for brands to get traffic on Facebook. What’s harder for them is finding ways to convert traffic into sales and activity, and I think the Timeline style profile with its huge pictures is something that they can really use to benefit themselves. If people didn’t like Timeline, they should just stop using Facebook and I doubt that will happen. That being said, while I think the big pictures on Timeline are beautiful, I think the Timeline interface isn’t perfect from a chronological perspective. The columns are awkward for scanning information and I think they would do well to revisit that somehow. I don’t begrudge Facebook for trying to make money with ads as long as they follow through on their promises not to give away personally identifying info to companies and generally create a pleasant service.
Yeah, that’s Google webfonts on Windows. I’ll poke around for a fix.
What is the source of the paraphrased quote, “the smartest people in the world are working hard to come up with ways to get you to click on ads.”?
It makes little sense to label a low-margin web business model “pathetic.” Facebook has a high bottom-line because of it, not in spite of it.
Interesting… For casual browsing, the timeline is kind of a fun way to explore someone’s profile, but I see what you mean as it being more difficult to avoid ads that show up in the timeline (when they do show up).
Facebook is trying to integrate brands much more intimately into our everyday experience, so perhaps rather than ads, they’ll just upsize any post or share that includes a paying brand (so a share via Coca Cola would be larger than a share from your grandma).
The web font fix is NOT to use the Google Webfonts font-face declaration. Download the font file and use font-squirrel to generate your own font-face files.
My understanding is that they’re coming online soon.
You said:
“With 2011 revenues of about $3.3 billion spread across 800,000,000 users, they’re making about $4 a user a year. That’s pathetic given how much the site is used.”
I like your blog and was having the pleasure of reading it until I read your statement that I quoted above. To my mind, with those statements, you failed to understand the business model of Google. The fact that, on average, millions of people are spending on google a lot less than they would ever care to spend is the whole beauty of the business model.
To maximize the average amount that people spend on google to the point that it starts to hurt will be detrimental to google.
I would prefer that Google optimize its income per person. So how does google do that? By doing experiments like Timeline.
Well if you use AdBlock on Firefox or in Chrome. Why bother about this. Theres always a solution…
This all takes me back to my political psychology coursework where all of the heuristic shortcut discussions left me frustrated by the level of activity given to figuring out how to frame issues to get voters to not bother (or to single-issue focus) rather than to engage as a whole. Maslow would be aghast.
@Loreto: I don’t agree. Given the obscene amount of data that facebook collects about us, they should be able to beautifully match ads to customers. If they can do this they’re wasting much less money on advertising by showing ads to fewer people, but people that are more likely to agree. This should increase the cost per impression or click, but also increase conversions drastically. Implied in your position is the assumption that the future of online advertising is about lots of microtransactions. It follows from that that there’s still a very low success rate per ad, otherwise the cost of customer acquisition will be unreasonably low.
@Haram: That’s fair, though when I use timid language I don’t get featured on HN :). I can’t decide if facebook is competing on scale/price, or competing on differentiation. Both of which have different impacts on the margins they can expect. The nature of massive quantities of data could let them do both, and without viable competitors it’s hard to guess where the margins will fall.
Typo in second to last paragraph: “This all reminds of of a recent”. (repeating “of”)
Yep, I started noticing Reddit advertisement only after I got hooked on Scrolldit.
@ENC, thanks, fixed!
Google’s revenue in 2004 was $3 B, with a net income margin of 13%. In 2010 it was $29 B, with a net income margin of 29%.
Facebook’s revenue today is about $3 B. Not sure on margin. So… 6 years out, will they be in a better or worse spot than Google is today?
To take the Kahmeman argument one step further, when you deplete System 2 with even trivial tasks people become more likely to rely on their System 1, which is for the most part very gullible as long as negative associations aren’t raised.
That is: people become more gullible at believing a neutral second task when you’ve just made them work to understand something. Seems to fit your argument.
@Max: Very interesting. It could be all about depleting S2 rather than accessing it. Again, still a conspiracy theory! :)
[...] making off me in advertising (i.e., as much as they can charge for my eyeballs). Jeff DeChambeau wrote today that, for Facebook, it comes out to about $4 per user per year. I’d pay [...]
Your line spacing is way too much, makes article hard to read.
[...] Archives January 18, 2012 Facebook’s Timeline could be intentionally difficult to use [...]
[...] Facebook Timeline is too awful to be an accident [...]
I multiplied 231 by 371 faster than my eyes dart to a Victoria Secret ad. Are my system 1 and system 2 reversed? Should i seek help?
interesting idea. I definitely can appreciate the psychologically devious ploy of tiring somebody out so they crash back into level 1 and make messy decisions.
however with timeline,
1. nobody ever goes there AFAICT. I never look at anybody’s timeline except on first adding them. the vast majority of the time is in the stream or looking at pages or messaging.
2. I never really notice ads on facebook, precisely because they don’t stand out. Google used to coach adsense users to use colors that stood out. Remember how glaring everybody was with the adsense blogs ? I had to go look at timeline to see if there really were ads there. There were. They were some small boxes with writing on them, but I’ve already forgotten what they said. I don’t even think my brain bothered to read them at all.
3. Facebook said straight out that it was about moving into a life product and trying to get people more personally invested in Facebook (my whole life is there) which of course makes people far less likely to leave. That would be its main purpose IMO. There’s also the possibility to get more data by asking for data about times prior to The Invention of Facebook and thus market ads better.
@gregg dourgarian: Humble brag? http://www.urbandictionary.com/define.php?term=Humble%20Brag
Jeff, I’ve recently started reading “Thinking, Fast and Slow” and I found your application of it to online advertising interesting.
By weaving ads (unexpected, to some degree) with information about your social life (expected), this is the online ad equivalent to the gorilla experiment referenced. It’s certainly a novel way of attempting to combat ad blindness. More often than not, I tend to be amused by the somewhat random nature of Facebook ads that I encounter but maybe this will prove to be a more effective channel.
I came across this post on your blog via Hacker News and I am very intrigued. I’ve been wrestling with this issue on a project we are currently working on in regards to member privacy versus a member supported platform. Although unrelated to the point of your post it nonetheless gives me some insight into the advertising model that is alien to me. I get the point of advertising but I rebel against it to the apparent detriment of my own projects. But my experience thus far is that most people seem to be happy to get free service instead of paying for it even if it means giving up some privacy. So FB might actually be right about privacy not being an issue.
Excellent article and I look forward to your future ones. It gave me some insight I had not considered. Thank you.
[...] Jeff Dechambeau, a Senior Analyst at T4G, an IT services company, made a comment on a start-up community site (Facebook timeline is too awful to be an accident) on how Facebook is manipulating user behaviour to increase brain activity and increase susceptibility to advertising , along with a link to his personal blog. [...]
[...] scanning someone’s Timeline. It’s a very unpleasant experience,” DeChambeau said in a blog post. “When information is organized in a list, it’s trivially easy to scan it, but with [...]
timeline looks great esp. compared to this site :P
not sure why the critique of timeline was mixed in with this interesting post about ad psychology.
i wonder to what degree ad placement actually went into the design of timeline- i don’t think I’ve seen an ad in timeline yet. guess it’s only a matter of time
Kahneman & Tversky focused mainly on cognitive tasks. The task you are describing (separating ads from content) is more like a scene-recognition task that can be done pre-attentively by people who would own a computer and use Facebook. There is little doubt that the decision to insert ads into the timeline was made to thwart our scene-recognition skills – and I’ve noticed an increase in designers and product managers who have some neuroscience background.
That said – I doubt this was driven by the designers or product managers. Slipping ads into areas where we expect to see content is becoming common practice. Google pioneered this with “sponsored” search results. Ebay was soon to follow with “featured” listings. Neilson Norman group have done numerous studies on the effects and Facebook is late to the table. The people advocating this sort of thing were probably just as keen to insert ads into the newsfeed. If tested – I would bet that ads in the newsfeed would have the same effect as ads in the timeline.
Facebook’s revenue is a morsel compared to what it will be, so I don’t know that they really care about 4$ / person if that is a valid calculation, and I don’t know that it is. I think they are still like the drug dealers who give away stuff for free cause they know that at a future date they’ll be laughing.
[...] 5. A conspiracy theory about how timeline is so bad. [...]
[...] (Hat tip) Advertisement LD_AddCustomAttr("AdOpt", "1"); LD_AddCustomAttr("Origin", "other"); LD_AddCustomAttr("theme_bg", "fff"); LD_AddCustomAttr("theme_border", "777"); LD_AddCustomAttr("theme_text", "222"); LD_AddCustomAttr("theme_link", "004276"); LD_AddCustomAttr("LangId", "1"); LD_AddCustomAttr("Autotag", "technology"); LD_AddSlot("wpcom_below_post"); LD_GetBids(); Like this:LikeBe the first to like this post. [...]
[...] France is building a theme park around Napoleon? I’m not sure having the Emperor come off as second fiddle to some parks featuring a cartoon [...]
[...] 5. A conspiracy theory about how timeline is so bad. [...]
[...] DeChambeau: Jeff DeChambeau – Behavioral economics and facebook conspiracy theories People recognize that advertising is about psychology, and the bible of psychological decision [...]
[...] Economics and Facebook Conspiracy Theories (( Jeff DeChambeau [...]
[...] ads, opening up one door into Timeline for branded content. On the heels of that came a fascinating blog post by business analyst Jeff DeChambeau arguing that Timeline is, in fact, explicitly designed for the benefit of advertisers. DeChambeau [...]
[...] of Instargam is id both more accessible and emotionally affective for marketing purposes. Deep visual brand integration within Timeline allows marketers to motivate viewers using strong peripheral appeals. The ‘value’ [...]
Join the discussion